Deflations in History
Even though experiences with falling prices have been rare during the postwar period, deflation was widespread during the 1930s, and recorded historical episodes extend back to ancient and medieval times. With deflation having resurfaced as a major policy concern in the years following the global financial crisis, this chapter compares the properties of the earlier deflations with more recent episodes in both Europe and Asia. In focusing upon the determinants of deflation, its impact upon the economy as a whole, and the role of monetary policy, we see that even though deflation remains, in essence, a monetary phenomenon, combatting its effects, remains far from straightforward. A striking feature of the twenty-first-century deflations, for example, has been the discrepancy between consumer and producer price movements seen after 2008 that occurred in conjunction with sharp declines in commodity prices. Policymakers both past and present have had to contend with a variety of downward pressures on the money supply as well as complications arising from supply shocks and other negative forces.
KeywordsDeflation Gold standard Monetary policy Price scissors Supply shocks
Hence followed a scarcity of money, a great shock being given to all credit, the current coin too, in consequence of the conviction of so many persons and the sale of their property, being locked up in the imperial treasury or the public exchequer … The facilities for selling were followed by a fall of prices, and the deeper a man was in debt, the more reluctantly did he part with his property, and many were utterly ruined … (Roman historian Tacitus’ description of the events in 33 AD, quoted in Temin 2013, pp. 141–142)
Experiences with falling prices have been rare since the 1930s. However, deflation not only has a long history going back to ancient and medieval times but also has resurfaced as a major policy concern in the years following the global financial crisis. This chapter compares the properties of the earlier deflations with more recent episodes in both Europe and Asia, focusing upon the determinants of the deflation, its impact upon the economy as whole, and the role of monetary policy. Co-movement between different price series cannot be taken for granted, and the post-2008 Chinese, Japanese, and Spanish cases have been marked by a large discrepancy between consumer and producer price movements. The role of supply-side factors, in particular the sharp oil price declines seen after the global financial crisis, must be considered in addition to purely monetary factors. Indeed, whereas earlier deflations like those of the 1930s occurred in the face of monetary contraction, the post-2008 experience suggests that even more expansionary policy cannot be counted on to offset downward pressures on the price level. The years since the global financial crisis have also been marked by a record disconnect between the stance of central bank policy and the rates of expansion achieved with regard to the actual amount of money in circulation. Although deflation remains, in essence, a monetary phenomenon, the historical record confirms that combatting its effects with monetary policy remains a tricky proposition. Policymakers both past and present have had to contend a variety of downward pressures on the money supply and complications arising from exchange rate constraints, supply shocks, and other negative forces.
In the days of commodity money, bullion shortages in fifteenth-century Europe were associated with deflation that ended only with “a new wave of devaluations combined with heavy emissions of … copper currency” (Day 1978, p. 47). The collapse of the Roman Empire represented another negative monetary shock as coinage supplies became dependent upon the Germanic tribes’ ability to operate the old Roman mints (Horesh 2014, p. 86). Similarly, the deflations of the late nineteenth century in the United Kingdom and United States occurred when growing scarcity of gold put downward pressure on the money supply and prices alike under the classical gold standard (Friedman 1992). The most well-known historical episode of deflation during the Great Depression of the 1930s also occurred in conjunction with monetary contraction, including the astonishing one-third drop in the US money supply between 1930 and 1933 (Friedman and Schwartz 1963). More recently, monetary tightening by the Bank of Japan in 1989 sets the stage for Japan’s stock market crash and ongoing “Lost Decade” and subsequent deflation concerns. Even briefer, and less well-known, deflationary episodes like that experienced by the Confederate States of America in early 1864 can be readily linked to monetary contraction. In this particular case, the April 1864 currency reform that reduced the money supply by one-third in the Eastern Confederacy induced a brief deflation that stood in stark contrast to the extensive inflationary trend both before and after this event (Burdekin and Weidenmier 2001).
In contrast to these earlier episodes of deflation, worldwide concerns with deflation emerged in the aftermath of the global financial crisis even in the face of substantial monetary easing. In the case of the United States, for example, the Federal Reserve’s policy response seemingly could not have been more different from the early 1930s case. The Federal Reserve quadrupled the size of the monetary base between 2008 and 2016 and, like many other central banks around the world, engaged in successive rounds of quantitative easing after the onset of the global financial crisis. However, even the maintenance of near zero interest rate targets by the Federal Reserve and European Central Bank was not enough to stop inflation falling close to zero in many countries and actually dropping into negative territory in such Eurozone economies as Greece and Spain. Deflation concerns also reemerged in both China and Japan notwithstanding expansionary central bank policy and currency depreciation. Indeed, some central banks, like Sweden’s Riksbank and the Swiss National Bank, were prompted to pursue not only quantitative easing but also negative interest rates. In Sweden’s case, this unprecedented move was in part a reaction to the strengthening of the Swedish krona against the euro following the European Central Bank’s own policy loosening (Spence 2015). This is in a small way reminiscent of the experience of the 1930s, when waves of competitive devaluations led to a “race to the bottom” after the abandonment of the international gold standard.
Besides the stance of monetary policy, another apparent contrast between earlier deflationary episodes and the post-2008 experience concerns the relationship between consumer and producer prices. Although not necessarily moving in lockstep, consumer price declines generally went hand in hand with declines in producer prices. A different element has emerged in the post-2008 Chinese case, however, with mild consumer price deflation being accompanied by substantially greater deflation in producer prices. This gap between consumer and producer price movements approached 8% in 2016, a phenomenon never present in any of the earlier deflation episodes. Moreover, growing gaps between consumer and producer prices were also seen in Japan and Spain after the global financial crisis. Again mild consumer price deflation was accompanied by much sharper declines in producer prices. The Spanish deflation began only in 2013 and, like the Greek case, reflects pressures associated with membership in the Eurozone and inability to enhance competitiveness via currency depreciation. Complicating the Greek situation, of course, is the additional role played by ongoing social unrest and other long-standing problems – leading Hetzel (2014, p. 275) to conclude: “Greece was dealt a bad hand [but] it has also played poorly the hand it was dealt.”
The Historical Record
Although the deflations of the 1930s and 1890s were themselves accompanied by monetary contraction, this did not so much reflect policy intent as the confines of the exchange rate constraint under the gold standard. In the late 1890s, gold became scarcer in the absence of new gold discoveries, and, for currencies linked to gold, greater purchasing power for gold could be achieved only via a decline in the prices of goods and services. The deflationary pressures did not ease until the invention of the cyanide process, and new gold discoveries in South Africa, provided for renewed expansion in the world’s gold supply (Friedman 1992). Meanwhile, in the early 1930s, central banks were largely handcuffed by the fact that unilateral monetary expansion would erode the balance of payments, triggering a gold outflow that would eventually render infeasible the existing parity with gold. Only after the abandonment of the gold standard, beginning with the United Kingdom in 1931 and later the United States in 1933, was there any real scope for sustained expansionary policy (c.f., Eichengreen 1992).
Even unilateral exchange rate commitment can have similarly restrictive effects, leading Argentina to abandon its fixed exchange rate, and float its currency, in December 2001 after being mired in a 4-year recession. At that time, Domingo Cavallo, an architect of Argentina’s currency board system that was enacted in 1990 to help quell hyperinflation, acknowledged the extent of the policy dilemma: “I always thought that stopping falling prices would be easy ... But I realize now that stopping hyperinflation was the easy part” (see Luhnow and Druckerman 2001, p. A6). In general, the most effective way to fight deflation is always going to be to keep it from starting or at least to keep it from lasting long enough to become entrenched in expectations. With reference to today’s Eurozone, Eichengreen and Temin (2013) foresaw that the weaker nations’ attempts to remain part of the single currency were likely to give rise to similarly dangerous constraints with austerity measures being forced onto already weakening economies. Spain’s post-2013 deflation is far from the only issue here, with major concerns surrounding the other members of the so-called “PIIGS” group (Greece, Ireland, Italy, and Portugal).
Even though evidence of a link between deflationary episodes and overall declines in economic activity may be less than clear-cut outside the Great Depression (Borio et al. 2015), one reason to fear deflation in consumer prices is that, if it is expected that such price declines will continue in the future, there is an incentive to delay purchases. This then leads to a further reduction in aggregate demand, putting further downward pressure on prices and suggesting that deflation could be at least partially self-sustaining. Reinforcing the contractionary effects of postponed consumption, the Mundell-Tobin effect implies that deflation encourages more money hoarding owing to rising real rates of interest. Mundell (1963) shows that, with the money rate of interest responding less than proportionately to innovations in inflation, even fully anticipated inflation would reduce real money balances and reduce wealth. The converse that deflation adds to the real value of money holdings and increases total real wealth is seen by Tobin (1965) as potentially having the destabilizing effect of diverting savings away from capital formation. Although the implied decline in capital’s share in total wealth, and associated rise in its yield, would eventually reverse this process, “they may do so too little and too late” (Tobin 1965, p. 684).
Fisher’s (1933) debt-deflation mechanism represents another potential threat, whereby deflation adds to real debt burdens, in turn triggering bankruptcies and further declines in economic activity. Although typically thought of in terms of the 1930s experience, a precursor appears to have been seen in ancient Rome in 33 AD. At that time, the collapse of a housing boom forced senators to sell land to raise money to pay off their creditors. This, in turn, not only caused land prices to fall even more but also raised debt burdens in relation to senators’ property values and exacerbated the decline in overall spending (Temin 2013, p. 142). The dependence on land as collateral itself naturally has parallels with Japan’s “Lost Decade,” with falling land values playing a key role in triggering defaults on loans after 1989.
Balance sheet effects can be another important factor (Bernanke 1983). When debtors who default forfeit their assets to banks, plunging prices of the forfeited assets then hurt bank balance sheets and threaten the solvency of the banking sector. If banks curtail lending as a result, firms dependent on bank credit will, in turn, face a credit crunch as seen in 2008. If left unchecked, this could easily lead to further production cutbacks, intensifying the deflationary spiral. Firms’ borrowing difficulties will be further exacerbated as declines in asset prices reduce the value of their loan collateral. In the US case, the unprecedented debt buildup during the 1920s was followed by a sudden shift from a stable price environment to a deflationary environment at the beginning of the 1930s. This meant that firms and households faced an unexpected rise in their debt service costs at the very time that their ability to fund their debt diminished. There is little doubt that the unanticipated nature of the rising real debt burden at the beginning of the 1930s fueled the operation of a debt-deflation process in the United States at this time (Fackler and Parker 2005).
Although Japan’s post-1990 problems never approached those experienced during the Great Depression, balance sheet and collateral concerns again emerged in the aftermath of the initial asset price collapse. The dramatic fall in land prices seen in Japan after 1989 was of particular concern because land formed the collateral for so many loans. This was followed by an ongoing series of net worth shocks for both financial intermediaries and entrepreneurs that lowered lending rates and investment activity alike. Hirakata et al. (2016) identify overwhelmingly negative values for net worth shocks in Japan that persist from the early 1990s through the aftermath of the global financial crisis. The failure of repeated attempts at expansionary monetary and fiscal policy to reverse the negative trends partly reflects the fact that Japanese banks were seeking to bolster their own balance sheets, which had been hurt not only by nonperforming loans but also by the sharp drop in the market value of their equity holdings. Indeed, the sharp drop in the Japanese money multiplier in the 1990s (Hutchison 2004) served as somewhat of a precursor for the massive drop in the US money multiplier after 2008. The banks’ reduced willingness to lend and to circulate the new money being created by the Bank of Japan meant that these funds were often simply held within the banking system, thereby doing little to fuel new spending and combat ongoing deflation.
Delays in bank recapitalizations, as well as the lack of major structural reforms, undoubtedly are among the factors accounting for the length of the downturn in Japan. Although positive technology shocks may have played some part in the downward pressure on prices (Saxonhouse 2005), the overall performance of the Japanese economy since the beginning of the “Lost Decade” implies that any “good deflation” was outweighed by the bad in this case. In this regard, Hoshi and Kashyap (2015) point to some worrying parallels with the post-global financial crisis situation in countries like France, Italy, and Spain. The emergence of declining inflation, and outright deflation in the case of Spain, has only added to the difficulties. With falling inflation adding to debt burdens, there is the very real possibility of a debt-deflation spiral emerging at the national government level. Indeed, some estimates suggest that, should inflation remain at 1% or lower, this could be enough to make Italian government debt unsustainable (Stewart 2013).
The situation may actually be even worse than it appears given that upward biases in conventional consumer indexes mean that a measured inflation rate of 1% or less could well be masking an actual deflationary trend – notwithstanding the improvements made in response to the recommendations of the Boskin Commission (Boskin 2005). With Spain having experienced declining consumer prices since 2013, this put the Spanish government’s already heavily extended fiscal position (as with other PIIGS countries like Greece) at substantial risk. Public debt not only rose by approximately 60% of GDP in the face of the global financial crisis but also continued to increase afterward (Munevar 2016), leaving Spain with the unattractive combination of one of the highest nominal debt burdens in the Eurozone plus outright deflation in both consumer and producer prices.
Comparative monetary trends, 2008–2016
Total monetary base growth from September 2008 to January 2016
Total bank money growth from September 2008 to January 2016
Overall money supply growth from September 2008 to January 2016
The rise in Chinese bank lending for the most part involved funds going to the government’s state-owned enterprises, with the surging real estate markets serving as another catalyst for loan expansion. Outside of China, bank lending stagnated and commercial banks largely sat on the new funds generated by central bank open market purchases. Hanke (2016) blames this phenomenon on overly zealous tightening of bank regulation and supervision, arguing that these policies have been “ultra-tight and procyclical.” Regulatory stringency remains an understandable reaction to the excesses of the earlier 2000s, however, and there are obvious parallels between the Dodd-Frank Act in the United States and the earlier imposition of the Glass-Steagall Act in 1933. There is also the question of whether bank lending has been held back not by excessive regulation but rather by continued weakness in the commercial banks own balance sheets. Although substantial recapitalization was undertaken in the United States, the European banks still seemed to stand in need of more recapitalization rather than being in a position to benefit from regulatory rollbacks alone. Bank lending aside, the data do call attention to the fact that base money issuance by the European Central Bank has lagged well behind its counterparts, however. The fact that its quantitative easing was roughly one-third the scale of the Federal Reserve’s may not be the only reason for relatively greater deflationary concerns in the Eurozone than in the United States, but this could hardly have helped matters.
The actual extent of deflation concerns should itself not be assessed on the basis of consumer prices alone. There is always the possibility of relative price changes wherein some components of the price level fall, while others keep rising. Moreover, even though discussions about the role of monetary policy tend to focus on the behavior of headline price indices, deflation in less-heralded areas such as producer prices should not be ignored given the potential for eventual transmission to consumer prices. Wage rigidities add to the dangers associated with any such price drop given that pushing the real wage above equilibrium must sooner or later give rise to unemployment. This concern can be contrasted with the old costs of production view that saw falling prices as a sign of lower production costs and a reflection of improvements in transportation and technology. In this vein, Dickey (1977) argues that the US deflation of 1869–1896 was primarily of the “good” variety since relative price changes, profit expectations, and bond yields all suggest that supply-side influences dominated price movements over demand-side effects. In analyzing both the Canadian and US experiences prior to World War I, Bordo and Redish (2004) argue for a more mixed picture, however. In particular, they reject any simple demarcation between “good” and “bad” deflation and attribute the price declines to a combination of both negative money supply shocks and positive supply shocks.
As a possible means of achieving price-level stability without the mechanical constraints incumbent under gold standard or fixed exchange rate arrangements, price-level targeting has been proposed. In 1898 Swedish economist Knut Wicksell made a pathbreaking case for price-level stabilization in his address to Sweden’s Nationalekonomiska Förening (reprinted in English in Wicksell 1958, pp. 67–89). Wicksell’s proposal actually relied upon interest-rate targeting and, as noted by Jonung (1979), could not have been implemented under the gold standard because adjusting interest rates to their noninflationary level would have required independent control of the money supply. After Sweden left the gold standard for good on September 27, 1931, however, “the norm that Wicksell presented at the turn of the century became, some thirty years later, the official foundation for Swedish monetary policy” (Jonung 1979, p. 468). Meanwhile, Irving Fisher (1913, 1920) argued in favor of a monetary regime based on a standard of commodity prices. Whereas price-level targeting could offer a means of signaling the policymaker’s commitment to lowering real interest rates even after nominal interest rates have already been cut to zero, the postwar era has seen central banks adopt inflation targeting instead Following its enshrinement in the Reserve Bank of New Zealand Act 1989, the Bank of Canada, the Bank of England, the European Central Bank, and many others adopted formal inflation targets – with the Federal Reserve joining the party in January 2012.
Although the case for price-level targeting over inflation targeting is seen as remaining model dependent, Hatcher and Minford (2016) emphasize that price-level targeting becomes more appealing in the face of deflationary shocks when central banks are confronted by the zero interest rate lower bound. In this case, expectations of higher than usual inflation aimed at recapturing the price-level target should “induce negative real interest rates which stimulate economic activity out of the recession, putting an end to deflation and ending lower bound episodes rapidly” (Hatcher and Minford 2016, p. 348). Although no central bank has yet made the switch to a price-level target, the Bank of Canada seriously considered the case for price-level targets in preparation for its 2011 policy agreement with the government (Kahn 2009). Interestingly, Ruge-Murcia (2014) finds that, unlike other inflation targeting central banks, the Bank of Canada’s policies over the 1996–2013 periods were already observationally equivalent to price-level targeting. Specifically, in the Canadian case, shocks to the price level over the sample period were reversed, and there was none of the “drift” evident in the other cases (comprising the central banks of Australia, New Zealand, Sweden, and the United Kingdom). There have also been calls for price-level targeting in the United States both before and after the 2012 announcement of an inflation target. San Francisco Federal Reserve Bank President John Williams presented the case for this policy change in May 2017, pointing to the advantage of allowing inflation to remain above the Federal Reserve’s existing 2% inflation target to make up for periods when it was too low (Williams 2017). Similar sentiments were previously expressed by Federal Reserve Bank of Chicago President Charles Evans in October 2010 (Evans 2010).
The potential benefit of price-level targeting in anchoring price expectations is supported by Fregert and Jonung’s (2004) account of how wages were able to adjust downward under the price-level target introduced after Sweden left the gold standard in September 1931 (see also Berg and Jonung 1999). The introduction of price-level targeting admittedly did not preclude continued uncertainty among policymakers and the public about the future course of the Swedish price level. Nevertheless, in Fisher’s (1934, p. 331) view, the ensuing stability of the internal purchasing power of the Swedish krona in conjunction with this new stabilization policy “stand out as among the remarkable facts of the depression.”
Fisher himself called for an adjustable commodity-based dollar that would automatically offset any upward or downward pressure on its purchasing power. Whenever prices fell below target by 1%, for example, the dollar value of the resource unit would be raised by 1% – at the same time automatically lowering the number of resource units in the dollar.
Burdekin et al. (2012) further suggest that a stripped-down form of Fisher’s commodity price standard could have helped Austria maintain price stability during the gold standard era. When increased scarcity of gold produced lower goods prices under the classical gold standard, Austrian data imply that investors saw silver prices moving on a one-to-one basis with the aggregate price level as Austria bought and sold gold on the London market to maintain the targeted exchange rate. This implies that a silver peg could potentially have allowed for inflation targeting with this one commodity used as a proxy for overall prices. More recently, while not involving any explicit price-targeting strategy, the 1934 Silver Purchase Act was a significant part of the more expansionary policies adopted after the US exit from the gold standard in 1933. The empirical importance of the silver effect detailed in Burdekin and Weidenmier (2009) serves as a reminder that government bond purchases are not the only means of achieving monetary expansion. And it is not just in the 1930s that commodity purchase programs may warrant further consideration if very low interest rates, and liquidity-constrained banks, call into question the effectiveness of more conventional policies.
Consumer Prices Versus Producer Prices
In China, a prior bout of short-lived consumer price deflation arose in the midst of extremely tight monetary policy aimed at combating an inflationary spike in 1993–1994. While renewed consumer price deflation in the aftermath of, first, the Asian financial crisis, and, second, the global financial crisis had also been only intermittent in nature, the post-2008 drop in producer prices was much more pronounced. Kohler (2016) actually argues that the problem was of China’s own making insofar as China’s massive stimulus program launched in November 2008 led to oversupply of steel and other raw materials. Irwin (2017) also points to low oil prices as a source of deflationary pressures, along with “more worldwide capacity for major commodities like steel and aluminium than there is demand, in part because of China’s sheltering of state-run enterprises from the vicissitudes of the marketplace.” There may actually be something of a vicious circle here in that excess supply in China puts downward pressure on global commodity prices, in turn causing China to be faced by imported deflation that only exacerbates the original problem (Wan 2015). Liu et al. (2017) also place blame on the November 2008 stimulus package, arguing that, in artificially bolstering the ratio of investment to consumption, this worsened structural imbalances in China’s economy that induced the rising price scissors. Liu et al. therefore see the problem as deeper than a simple overcapacity issue and argue that the only way to eliminate the price scissors is through structural reform on the demand side that would boost PPI by encouraging greater private investment.
Components of Chinese total factor productivity, 2001–2013
Technical efficiency change
Economies of scale
Efficiency in resource allocation
Total factor productivity growth
The recent discrepancy between producer and consumer prices echoes a phenomenon that existed in China before price reform began in the 1980s. The controlled prices under the command economy had “allowed a form of price scissors to open by allowing raw materials prices, including agricultural prices, to increase more slowly than consumer prices” (Wedeman 2003, p. 49). The resulting resource transfer from agricultural areas to the urban and coastal areas itself had parallels with Preobrazhensky’s theory of primitive socialist accumulation in the former Soviet Union, whereby resource transfers from the agrarian sector were to fund growth in the industrial sector (see also Burdekin 1989). It is ironic that a similar price scissors, previously thought to be merely a relic of the old command economy, reemerged in the predominantly market-based system in place today. Moreover, some degree of analogy with the old price scissors remains insofar as expansion of the services sector at the expense of the agricultural sector has helped boost consumer prices relative to producer prices in the new millennium (Lu and Wang 2015). The differential between consumer and producer prices has itself been found to be linked with stock market performance in China, and Guo and Liu (2015) find evidence of a long-run relationship between the CPI-PPI differential and the Shanghai A-share index over the 2001–2011 period. They attribute this linkage to the implications of growing CPI-PPI differentials implying larger profit margins and stronger company performance – albeit without taking into account the role of policy-driven factors like bank deposits and bank loans (Liang and Willett 2015).
Guo and Liu’s (2015) emphasis on the price scissors’ impact on profit margins points to the phenomenon potentially having at least some positive effects. And the fact that the Shanghai market continued to rise even as PPI deflation accelerated after 2010 is not inconsistent with Guo and Liu’s premise. On the other hand, there is the risk that declining producer prices will eventually pull consumer prices down as well, leading to the traditional concerns of consumers putting off nonessential purchases and wages needing to have downward flexibility in order to avoid the risk of workers being priced out of a job. Indeed, Zou (2016) argues that the scale of the recent producer price declines could lead to deflation becoming entrenched in China and potentially unleash a debt-deflation cycle. This is itself far from just a Chinese concern as similar price scissors developed in other nations facing deflationary pressures after the global financial crisis. Indeed, the price scissors reached 4–5% in Japan over the 2009–2016 period and also approached 5% in Spain, steadily increasing through 2016 after deflation in consumer prices set in during 2013.
Traditionally, it would indeed be expected that, even if consumer prices initially remained reasonably stable, sharp producer price declines would sooner or later translate into significant consumer price deflation. This is consistent with the unidirectional causality from producer prices to consumer prices identified in such studies as Caporale et al. (2002), who find this pattern to consistently apply across all the G7 countries over the 1976–1999 period. Some controversy surrounds the question of whether this standard direction of causality has held in the Chinese case, however. Fan et al. (2009), for example, instead find evidence of causality running from consumer to producer prices over the 2001–2008 period in China. Although they attribute this to a dominant role of demand-side factors over supply-side factors during this period, their sample predates the more recent oversupply concerns that emerged after 2008.
Japan and Spain represent the two largest economies, other than China, to have so far experienced outright deflation in the aftermath of the global financial crisis. As with the Chinese situation, structural concerns apply in each case. However, worries about underinvestment in Japan contrast with the overinvestment that was a hallmark of China following the launch of the massive November 2008 stimulus package. Meanwhile, Spain’s delays in bank recapitalization have been compared to Japan’s own inadequate policy responses as it entered its first “Lost Decade” in 1990 (Hoshi and Kashyap 2015). Although it may seem that Spain’s mildly positive growth since deflation began in 2013 implies a modern-day instance of “good” deflation, Munevar (2016) points to not only large-scale job losses but also to so many people losing their homes that Spain alone accounted for nearly 30% of the entire Europe-wide stock of empty houses by 2016. Moreover, Spain’s limited GDP growth has relied upon highly expansionary fiscal policy that is likely non-sustainable with continued deflation pushing real debt burdens up.
Notwithstanding the different structural challenges facing the Chinese, Japanese, and Spanish economies after the global financial crisis, one deflationary catalyst they all faced was the worldwide decline in commodity prices since 2008, highlighted by a more than 70% drop in crude oil prices. Insofar as this affected producer prices more than consumer prices in Japan and Spain, like in China, we have a potential common element giving rise to the price scissors phenomenon in three otherwise quite different economies. Short-run effects of oil price changes are generally confined to such oil-related consumption areas as petrol and heating oil. The initial impact on non-oil consumption will be constrained by the time required for increases in firms’ input costs to be passed through – and, in the case of Spain, Álvarez et al. (2017) stress the slow speed and limited strength of this mechanism. Similarly limited transmission between oil price movements and consumer prices in the Eurozone as a whole is suggested by Castro et al. (2016), who conclude that deflation in consumer prices is unlikely to ensue from oil price declines alone. Limited effects on consumer prices, combined with more substantial and direct effects on producer prices, would, however, be very much in line with the widening gap between consumer prices and producer prices that emerged when oil prices plummeted after 2008.
Burdekin and Hu’s (2018) vector autoregression (VAR) analysis of post-2000 data from China, Japan, and Spain offers some support for a supply-side, commodity-based impetus to the recent deflationary pressures insofar as oil price movements (proxied by West Texas Intermediate crude) exert effects on PPI that are significant at the 99% confidence level for all three countries. Oil prices also significantly Granger-cause the CPI in Japan and Spain at the 99% confidence level. Meanwhile, weaker oil price effects on Chinese CPI help explain the more dramatic price scissors in China than in the other two countries – with lower commodity prices seemingly driving down both price series in the other countries but definitively impacting only PPI in China. Oil price shocks were previously found to play a similarly important role in determining Chinese producer price movements over the 1996–2009 period by Cai and Wang (2012). Cai and Wang’s structural VAR analysis identified significant effects of global commodity prices on China’s PPI, whereas there was no evidence of any such transmission to China’s CPI. Tang et al. (2014) find further evidence of global commodity price movements being transmitted to Chinese PPI but not CPI using data from the 2000 to 2011 period. Although Tang et al. (2014) point to profits in downstream industries accordingly being squeezed when commodity prices are rising, the other obvious implication is that falling commodity prices should be profitable for such industries, at least in the short-run – in line also with Guo and Liu’s (2015) finding of positive stock market effects arising from any widening of the price scissors.
The danger of a widened price scissors, however, is that falling PPI eventually gives rise to larger CPI declines, in turn raising fears of negative effects on consumption, potential debt-deflation, and also pressure for wage cuts. Analysis of the price scissors itself in Burdekin and Hu (2018) shows the lagged change in oil prices to be significant at the 99% confidence level for China and Japan and at the 95% confidence level for Spain, with the expected negative sign. Thus, falling commodity prices appear to lower PPI more than CPI, hence exacerbating the extent of the price scissors. Insofar as supply-side deflationary pressures have indeed outweighed the effects of expansionary demand-side policies in China, Japan, and Spain, it has to be of considerable concern that the extensive expansionary policy initiatives undertaken since 2008 apparently proved insufficient to reign in deflationary pressures.
An Alternative Commodity-Based Standard?
To the extent that post-2008 deflationary pressures can indeed be considered a supply-side phenomenon, the price scissors developing in countries like China, Japan, and Spain would simply be signaling how such effects initially manifest themselves more in PPI than in CPI. This raises two rather troubling implications, however. First of all, even massive monetary stimulus seems to have been insufficient to ward off ongoing deflationary pressures since the global financial crisis. Second, it seems highly likely, if not inevitable, that sharply falling PPI will sooner or later translate into significant consumer price deflation if left unchecked. In light of this, might central banks consider striking at the root of the problem and target commodity prices directly? Although a formal commodity-based standard may not be a likely contender today, serious consideration was given after World War II to buffer stocks of raw materials being incorporated under the new International Monetary System (see Burdekin 2007). In a more limited application of the idea of commodity-backing, Frankel (2003) and Frankel and Saiki (2002) suggest that policymakers in commodity-dependent nations should tie monetary policy to the nation’s primary commodity price. That is, stimulate monetary expansion whenever the nation is faced by declining demand for the commodity (or commodities) in question. Meanwhile, less specialized producers could enjoy analogous benefits by fixing the price of a basket of export commodities in terms of local currency (Frankel 2005).
Although the case for pegging the export price will generally be stronger when exports are large relative to total GDP, bigger economies could still benefit from such a commodity-based strategy, at least in comparison to the more popular alternative of pegging to the US dollar. For example, had Argentina adopted a wheat-peg rather than a dollar-peg during 1991–2001, the Argentinean peso would have been pushed down by falling commodity prices in contrast to the sharp appreciation actually seen over the latter part of this period (Frankel 2003). The dollar peg was deflationary for Argentina just as the gold peg was deflationary for the United States and many other countries in the late nineteenth century and during the Great Depression – and the authorities in each case ended up maintaining a fixed single-asset price, while almost all other prices declined. For Latin American countries as a whole, Frankel (2011) stresses the advantage of targeting PPI rather than CPI as a basis for monetary stabilization policy. Frankel demonstrates how, under this policy, falling commodity export prices would, for example, automatically trigger an expansionary response and currency depreciation – whereas CPI targeting would not. Meanwhile, in the case of rising import prices, CPI targeting would induce the central bank to tighten monetary policy and appreciate the currency to offset the rise in local currency prices, thereby actually contracting in the face of a negative shock.
Tying monetary expansion to commodity prices rather than consumer prices alone certainly seems like a valid proposal today for any countries facing expanding price scissors of the type seen in China, Japan, and Spain – regardless of whether the economies concerned are necessarily as commodity-dependent as those of Latin America. Indeed, under inflation targeting, the existing policy focus is even narrower in that the target is typically set in terms of a “core inflation” measure. Core inflation is intended to represent the long-term trend in prices, which generally involves excluding food and energy prices on account of their greater volatility. The problem is that, under this strategy, even sustained movements in oil prices such as those seen in the aftermath of the global financial crisis will fail to elicit any policy response. Quite aside from the question of price level versus inflation targeting, a broader measure of the policy target seems warranted in the post-global financial crisis world.
Deflation: Always and Everywhere a Monetary Phenomenon?
In essence, declining prices must imply insufficient monetary expansion, i.e., too little money chasing too many goods. This premise seems to receive ample support from the deflationary experiences of the nineteenth century and 1930s. Each episode was marked by clear declines in the money supply and a largely passive policy, limited at the time by the constraints of the international gold standard. On the other hand, post-2008 deflationary pressures have emerged despite record rates of base money expansion around the world and central banks pushing interest rates down close to zero or even beyond. Although it may not seem reasonable to label such herculean efforts as being insufficient, the reality is that the expansionary push has been offset by other factors reversing the positive effects on the price level. The first has been a decline in the money multiplier and cutback in bank lending, dwarfing in size the movements in the same direction that were seen during the 1930s – as well as in Japan in the 1990s. The second key consideration concerns how the much more limited expansion in overall money supply has seemingly been offset by supply-side effects like the sharp declines in the price of oil and other commodities after the global financial crisis. More worrying still, these effects were initially felt primarily in producer prices rather than consumer prices in countries like China, Japan, and Spain.
Post-2008 deflationary concerns have already drawn renewed attention to the case for major structural reforms in China, Europe, and Japan. Some old remedies may warrant more attention as well, however. Price-level targeting would require central banks to reverse any price declines that have already occurred, thereby potentially providing a more solid anchor for expectation formation and wage setting as seen in the Swedish example of the 1930s. At the same time, the price scissors phenomenon, under which producer price declines can, in the short run, greatly outstrip the fall in consumer prices, warrants policymakers’ attention too. Giving at least some weight to commodity prices in the design and implementation of monetary stabilization may be appropriate, again building upon proposals and theories that arose during deflationary episodes of the past. In this respect, it is worth bearing in mind the old adage: “Those who cannot remember the past are condemned to repeat it.”
- Álvarez LJ, Sánchez I, Urtasun A (2017) The effect of oil price fluctuations on Spanish inflation. Banco de España, Madrid, May 11. https://www.bde.es/f/webbde/SES/Secciones/Publicaciones/InformesBoletinesRevistas/ArticulosAnaliticos/2017/T2/files/beaa1702-art14e.pdf
- Andersson K, Berg C (1995) The inflation target in Sweden. In: Haldane AG (ed) Targeting inflation: a conference of central banks on the use of inflation targets organized by the Bank of England. Bank of England, London, pp 207–225Google Scholar
- Bernanke BS (1983) Non-monetary effects of the financial crisis and the propagation of the Great Depression. Am Econ Rev 73:257–276Google Scholar
- Borio C, Erdem M, Filardo A, Hoffman B (2015) The costs of deflation: a historical perspective. BIS Q Rev 31–54Google Scholar
- Burdekin RCK (2007) Nontraditional monetary policy options and commodity-based stabilization policy. Int Econ Finance J 2:1–18Google Scholar
- Burdekin RCK, Hu X (2018) Deflationary pressures today: the Chinese, Japanese and Spanish cases compared. Chin Econ, forthcomingGoogle Scholar
- Burdekin RCK, Weidenmier MD (2009) The development of ‘non-traditional’ open market operations: lessons from FDR’s silver purchase program. In: Atack J, Neal L (eds) The origins and development of financial markets and institutions: from the seventeenth century to the present. Cambridge University Press, New York, pp 319–344Google Scholar
- Cai Hong-bo, Wang Jun-hai (2012) International commodity price and China’s economic fluctuation: empirical analysis based on different crude oil price Shocks [in Chinese]. Int Bus (Univ Int Bus Econ) [国际商务(对外经济贸易大学学报)] (1):54–64Google Scholar
- Dickey GE (1977) Money, prices and growth: the American experience 1869–1896. Arno Press, New YorkGoogle Scholar
- Eichengreen B (1992) Golden fetters: the gold standard and the Great Depression, 1919–1939. Oxford University Press, New YorkGoogle Scholar
- Evans CL (2010) Monetary policy in a low-inflation environment: developing a state-contingent price-level target. Remarks delivered before the Federal Reserve Bank of Boston’s 55th economic conference on October 16. http://www.chicagofed.org/webpages/publications/speeches/2010/10_16_boston_speech.cfm
- Fisher I (1920) Stabilizing the dollar. Macmillan, New YorkGoogle Scholar
- Fisher I (1934) Stable money: a history of the movement. Adelphi, New YorkGoogle Scholar
- Frankel JA (2011) Comparison of product price targeting and other monetary anchor options for commodity exporters in Latin America. Economia 12:1–57Google Scholar
- Fregert K, Jonung L (2004) Deflation dynamics in Sweden: perceptions, expectations, and adjustment during the deflations of 1921–1923 and 1931–1933. In: Burdekin RCK, Siklos PL (eds) Deflation: current and historical perspectives. Cambridge University Press, New York, pp 91–128CrossRefGoogle Scholar
- Friedman M (1992) Money mischief: episodes in monetary history. Harcourt Brace Jovanovich, New YorkGoogle Scholar
- Friedman M, Schwartz AJ (1963) A monetary history of the United States, 1857–1960. Princeton University Press, PrincetonGoogle Scholar
- Ge Feng [葛丰] (2017, March 20) PPI, CPI price scissor enlarged, adding difficulty to macroeconomic control [in Chinese]. China Economy Weekly [中国经济周刊], No. 11. http://www.ceweekly.cn/2017/0320/184609.shtml
- Guo Sanshua [郭三化], Liu Jun [刘俊] (2015) Empirical study of the relationship between the CPI-PPI differential and the stock index [in Chinese]. J Financ Dev Res [金融发展研究] (7):72–76Google Scholar
- Hamilton JD (1992) Was the deflation during the Great Depression anticipated? Evidence from the commodity futures market. Am Econ Rev 82:157–178Google Scholar
- Hanke SH (2016) State money and bank money: lifting the fog around QE. Cent Bank, June 6. https://www.cato.org/publications/commentary/state-money-bank-money-lifting-fog-aroundqe
- Hatcher M, Minford P (2014) Inflation targeting vs. price-level targeting: a new survey of theory and empirics. VOX, CEPR’s Policy Portal, May. http://voxeu.org/article/inflation-targeting-vs-price-level-targeting
- Hetzel RL (2014) Should Greece remain in the Eurozone? Econ Q 100(Third Quarter):241–278. Federal Reserve Bank of RichmondGoogle Scholar
- Horesh N (2014) Chinese money in global context: historic junctures between 600 BCE and 2012. Stanford University Press, StanfordGoogle Scholar
- Irwin N (2017, April 26) The low-inflation world may be sticking around longer than expected. New York Times. https://www.nytimes.com/2017/04/26/upshot/the-low-inflation-world-may-be-sticking-around-longer-than-expected.html
- Kahn GA (2009) Beyond inflation targeting: should central banks target the price level? Econ Rev (Third Quarter):35–64. Federal Reserve Bank of Kansas CityGoogle Scholar
- Kohler A (2016) The disaster of negative rates. The Australian, April 12. http://www.theaustralian.com.au/business/opinion/alan-kohler/the-disaster-of-negative-rates/news-story/9d7f93a8a0e1188dff29b31b4d70f372
- Liu Fengliang, Zhang Xiaomeng, Yu Ze (2017) High investment, structural imbalance, and the CPI & PPI deviation [in Chinese]. J Financ Res [金融研究] (2):54–69Google Scholar
- Lu Jie, Wang Gaowang (2015) A structural explanation on the ‘divergence’ between the CPI and PPI [in Chinese]. Jingji Yanjiu/Econ Res J 50(4):136–149Google Scholar
- Luhnow D, Druckerman P (2001, December 4) Long hailed as hero, reformer in Argentina sees his dream sour. Wall Street Journal, pp A1, A6Google Scholar
- Munevar D (2016) Spain: the politics of austerity and deflation. Institute for New Economic Thinking, July 4. https://www.ineteconomics.org/perspectives/blog/spain-the-politics-of-austerity-and-deflation
- Ruge-Murcia F (2014) Do inflation targeting central banks implicitly target the price level? Int J Cent Bank 10:301–326Google Scholar
- Spence P (2015, September 27) How Sweden’s negative interest rate experiment has turned economics on its head. The Telegraph. http://www.telegraph.co.uk/finance/economics/11895084/How-Swedens-negative-interest-rates-experiment-has-turned-economics-on-its-head.html
- Stewart H (2013, November 13) Deflation fears stalk Eurozone as Spain reports fall in prices. The Guardian. https://www.theguardian.com/business/2013/nov/13/deflation-fears-eurozone-spain-fall-prices
- Temin P (2013) The Roman market economy. Princeton University Press, PrincetonGoogle Scholar
- Wan Zhao [万钊] (2015, August 10) Chinese PPI-CPI scissor enlarged, road ahead for structural adjustment will be long [in Chinese]. Reuters. http://cn.reuters.com/article/cn-ppi-colum-idCNKCS0QG0J420150811
- Wicksell K (1958) In: Lindahl E (ed) Selected papers on economic theory. Harvard University Press, Cambridge, MAGoogle Scholar
- Williams JC (2017) Preparing for the next storm: reassessing frameworks and strategies in a low R-star world. FRBSF Economic Letter 2017-13, May 8Google Scholar
- Wu Qiong [邬琼] (2016) Computation and analysis of Chinese factor productivity [in Chinese]. State Information Center [国家信息中心], 2 Sept. http://www.sic.gov.cn/News/455/6841.htm
- Zou Jingxian (2016) CPI, PPI diversion and deflation [in Chinese]. Int Econ Rev [国际经济评论] (4):129–139Google Scholar